CA
Listed on Cognition AI’s own careers site. You apply with them directly — we never stand between you and the employer.
What this role is
This role leads safety and alignment work for Devin, an autonomous AI agent that writes code and takes actions in production systems. You'll design evaluations, red-team the agent, develop alignment methods, and partner with the agent team to ensure safe behavior at scale—a hands-on research engineering position where safety work directly influences a deployed product.
Our summary, not Cognition AI’s wording. The full posting is on their site.
What they ask for
Required
- Hands-on experience in safety, alignment, evaluations, red-teaming, or interpretability at a frontier AI lab or published research
- Understanding of how agents fail in practice and ability to measure or reason about failure modes
- Proficiency in Python and PyTorch or JAX
- Ability to build evaluation infrastructure and run experiments at scale
- Empirical rigor in experiment design and honest reporting
Nice to have
- PhD in Computer Science, Machine Learning, Physics, Mathematics, or related discipline
- Prior experience at a frontier AI lab, applied AI company, or developer tools company
- Adversarial thinking and systems-breaking experience
- Published safety or alignment research